Electromagnetic valve fault diagnosis method and device based on multi-sensor data fusion

The invention discloses an electromagnetic valve fault diagnosis method and device based on multi-sensor data fusion, and the method employs a current signal and an air pressure signal as the characteristic signals of the electromagnetic valve fault diagnosis, carries out the further subdivision of the electromagnetic valve fault diagnosis, employs a wavelet packet accumulation EMD method to process the characteristic signals, and achieves the fault diagnosis of the electromagnetic valve. The EMD mode aliasing phenomenon is avoided, and the BP neural network is adopted as a fault diagnosis model to effectively diagnose the electromagnetic valve. In order to facilitate implementation of the fault diagnosis method, a device matched with the fault diagnosis method is further designed. The device can extract an air pressure signal and a current signal when the tested electromagnetic valve works, carries out fault diagnosis, can output adjustable fluctuating pressure, and simulates an air pressure sudden change en